search

rmcantin / bayesopt

BayesOpt: A toolbox for bayesian optimization, experimental design and stochastic bandits.
GNU Affero General Public License v3.0
396 stars 92 forks source link

bayesopt.optimize_discrete #9

Closed safooray closed 7 years ago

safooray commented 8 years ago

BayesOpt keeps trying the same point. Why does it do it when it already knows the outcome of that point? After a few iterations it forces one random sampling, but gets right back to the old point and gets stuck.

rmcantin commented 8 years ago

BayesOpt always assume that the function is noisy. Also, it might imply that your model is not suitable for your problem and have to tune the parameters.

Note that BayesOpt can work with discrete spaces, but it's better/easier for continuous.